Iron and Ferritin Modulate MHC Class I Expression and NK Cell Recognition

Abstract

The ability of pathogens to sequester iron from their host cells and proteins affects their virulence. Moreover, iron is required for various innate host defense mechanisms as well as for acquired immune responses. Therefore, intracellular iron concentration may influence the interplay between pathogens and immune system. Here, we investigated whether changes in iron concentrations and intracellular ferritin heavy chain (FTH) abundance may modulate the expression of Major Histocompatibility Complex molecules (MHC), and susceptibility to Natural Killer (NK) cell cytotoxicity. FTH downregulation, either by shRNA transfection or iron chelation, led to MHC surface reduction in primary cancer cells and macrophages. On the contrary, mouse embryonic fibroblasts (MEFs) from NCOA4 null mice accumulated FTH for ferritinophagy impairment and displayed MHC class I cell surface overexpression. Low iron concentration, but not FTH, interfered with IFN-γ receptor signaling, preventing the increase of MHC-class I molecules on the membrane by obstructing STAT1 phosphorylation and nuclear translocation. Finally, iron depletion and FTH downregulation increased the target susceptibility of both primary cancer cells and macrophages to NK cell recognition. In conclusion, the reduction of iron and FTH may influence the expression of MHC class I molecules leading to NK cells activation.

Keywords: HLA; IFNγ; MHC-I; NK cells; STAT1; iron.

Figure 1

Ferritin Heavy Chain (FHC) modulates the expression of MHC-class I molecules. (A) Mel30 and Mel35 primary melanoma cells where treated overnight with DFO and classical (HLA A/B/C) MHC class I surface expression was measured by flow cytometry. The dashed curve in the two histograms represents the isotype control; the white curve represents the untreated (Nt) control cells and the filled gray curve represents cells treated with DFO. Columns show statistical analysis of three consecutive independent experiments. P-values were calculated using paired Student t-test (^*p < 0.05; ^**p < 0.01; ^***p < 0.001). (B) Western blotting analysis of Ferritin Heavy Chain (FTH1) and Iron binding protein 2 (IRP2) in HeLa cell lines stably transfected with shCTRL or shFTH. Three different clones were selected with puromycin (2.5 ug/ml). Tubulin was used as loading control. (C) Flow cytometry measurements of labile iron pool (LIP) in HeLa shCTRL and two different shFTH clones. Cells were treated with Calcein-AM, followed by 1 h incubation with or without DFO. The ratio in the mean fluorescence intensity (MFI) between DFO-treated and untreated cells of three different experiment was measured and is presented in the histogram. P-values were calculated using 2-tailed Unpaired Student t-test. ^*p < 0.05, ^**p < 0.01. (D) Representative Flow cytometry analysis of classical (HLA A/B/C) surface expression in HeLa shCTRL and two different shFTH clones. (E) Relative expression of MHC-I mRNA in HeLa shCTRL and two different shFTH clones with respect to the housekeeping gene GAPDH. (F) Western blotting analysis of MHC-I (A–C) in HeLa cells stably transfected shCTR or shFTH.

Figure 2

In vitro and ex vivo modulation of ferritin heavy chain and MHC-I expression in mouse. (A) Immortalized Embryonic fibroblasts derived from NCOA4 WT and KO mice (MEFs) were surface stained for MHC class I molecules H2 K (b) (apex) and H2 D (b) (apex) and analyzed by flow cytometry. Data are representative of three independent experiments. (B) Cell surface levels of H2-K (b) (apex) and H2-D (b) (apex) were analyzed on freshly explanted splenocytes of NCOA4 KO mice compared to those of wt mice. White column represents wt (B6) mice while gray column corresponds to NCOA4 KO mice. Data are representative of three consecutive experiments. P-values were calculated using 2-tailed Unpaired Student t-test. ^*p < 0.05; ^** p < 0.01; ^*** p < 0.001.

Figure 3

Interferon-γ stimulation is affected by iron levels. (A) Mel30 and Mel-35 primary melanoma cell were grown in presence of DFO, IFN-γ, or a combination of both. Cells were stained for MHC class I molecule (HLA-A/B/C), and analyzed by flow cytometry. The dashed curve in the histograms represents the isotype control; the white curve represents the untreated control cells; the light gray curve represents the treatment with DFO; the black curve represents cells stimulated with IFN-γ, and the dark gray curve represents cells treated with DFO + IFN-γ. Columns show statistical analysis of three independent experiments. Statistical analysis was performed by ANOVA followed by Holm-Sidak's multiple comparisons test. ^*p < 0.05; ^**p < 0.01; ^***p < 0.001. (B) Representative flow cytometry analysis of classical (HLA-A/B/C) surface expression in HeLa shCTR and shFTH clone 1, grown for 12 h with or without DFO (100 uM) and then stimulated for 24 h with IFN-γ (20 ng/ml). Statistical analysis was performed using Student t-test comparing untreated to IFN- γ-treated cells, and comparing IFN-γ to IFN-γ /DFO treated cells. ^*p < 0.05, ^**p < 0.01 (C). HeLa shCTR and shFTH clone 1 were treated as described in Figure 3B, then total RNA was extracted, retro-transcribed into cDNA and the expression of MHC-I mRNA was analyzed compared to the housekeeping gene GAPDH. Statistical analysis was performed using Student's t-test comparing untreated to IFN-γ-treated cells. (D) Western blotting analysis of the indicated antibodies in HeLa shCTR and HeLa shFTH clone 1 cells treated as in Figure 3B. pSTAT1 (Y701) antibody was used to monitor STAT1 activity and tubulin was used as loading control.

Figure 4

Iron and Ferritin Heavy Chain reduction is associated with NK cell susceptibility of established tumor cell lines and primary melanoma cells. (A,B) Representative cytotoxicity experiments in which two primary melanoma cell lines, Mel-30 and Mel-35, where treated overnight with DFO. Curves show the NK cell recognition of DFO treated cells (gray squares), and untreated cells (white squares). Columns show statistical analysis of three consecutive experiments at the effector:target ratio of 25:1 and 12:1, respectively. (C,D) Representative cytotoxicity experiments in which MM07m shFTH and MCF7 shFTH (gray squares), or scramble (white squares) were tested for their susceptibility to NK cell killing. Column show statistical analysis of five consecutive experiments at the effector:target ratio of 25:1 and 12:1, respectively. P-values were calculated using paired Student t-test (^*p < 0.05; ^**p < 0.01; ^***p < 0.001).